首页|多目标下改进NSGA-Ⅱ的机械臂轨迹规划

多目标下改进NSGA-Ⅱ的机械臂轨迹规划

扫码查看
以UR5e机械臂为研究对象,针对机械臂作业过程的运动轨迹进行轨迹规划,机械臂关节空间轨迹采用七次B样条曲线构造,针对其需满足时间短、能耗低、运动平滑的要求采用非支配排序遗传算法(NSGA-Ⅱ),并针对其易陷入局部最优、收敛速度慢等缺点,采用佳点集和非均匀变异算子使初始种群分布更加均匀、加快收敛.仿真结果表明,七次B样条曲线所构造的机械臂轨迹速度、加速度、加加速度连续平滑且无断层现象,在运动学约束的前提下,改进后的NSGA-Ⅱ算法世代距离平均下降了76.96%,空间分布性指标平均下降了23.06%,其在收敛性、分布性上均优于原算法,更接近于Pareto最优解,有效实现了机械臂的多目标优化问题.
Trajectory Planning of Manipulator Based on NSGA-Ⅱ Under Multi-Objectives
Taking UR5e manipulator as the research object,trajectory planning is carried out for the motion trajectory of the manipulator during its operation.The joint space trajectory of the manipulator is constructed by cubic B-spline curve,and the non-dominated sorting genetic algorithm-Ⅱ(NSGA-Ⅱ)is adopted to meet the requirements of short time,low energy consumption and smooth motion.In view of its shortcom-ings such as easy falling into local optimum and slow convergence,good point set and non-uniform muta-tion operator are adopted to make the initial population distribution more uniform and accelerate conver-gence.The simulation results show that the trajectory speed,acceleration and jerk of the manipulator con-structed by the cubic B-spline curve are continuous and smooth,and there is no fault phenomenon.Under the premise of kinematic constraints,the generation distance of the improved NSGA-Ⅱalgorithm is reduced by 76.96%on average,and the spatial distribution index is reduced by 23.06%on average,which is better than the original algorithm in convergence and distribution,closer to Pareto optimal solution,and effectively realizes the multi-objective optimization problem of the manipulator.

trajectory planningtrajectory optimizationgenetic algorithmmulti-objective optimizationNSGA-Ⅱ algorithm

张禹、邸贺彤、陈志远

展开 >

沈阳工业大学机械工程学院,沈阳 110870

沈阳市政集团有限公司,沈阳 110026

轨迹规划 轨迹优化 遗传算法 多目标优化 NSGA-Ⅱ算法

2024

组合机床与自动化加工技术
大连组合机床研究所 中国机械工程学会生产工程分会

组合机床与自动化加工技术

CSTPCD北大核心
影响因子:0.671
ISSN:1001-2265
年,卷(期):2024.(5)
  • 6